AResearch Proposal:

SmellAbilities in Individuals with Alzheimer's Disease andAnosmia

Sara Penn

Stephen F. Austin StateUniversity

"The sensation and perception of smell are largely dependent onsniffing" (Desmond et al., 1998). Research has shown that sniffing,with or without an odorant present, can cause activation in theolfactory bulb (Desmond et al. 1998). Further, they discovered thatsniffing (whether or not an odor is present) stimulates activity inthe piriform cortex of the temporal lobe and that actually smellingan odor stimulates the temporal lobe as well as the frontal lobe. InUSA Today, John Gabrielli stated that there are two different aspectsto the olfactory process. One aspect is called the exploratoryphase-the sniffing, and the other is called the evaluative phase-thesmelling (1998, December). These pieces of information can be appliedto examine whether there is more olfactory activation in anindividual with anosmia versus an individual with Alzheimer'sdisease. Ongoing research has been conducted on these two diseasesfor years but there is not a lot of research comparing the twotogether. It has been discovered that many patients with Alzheimer'sdisease also have a decreased sense of smell (Andrew et al., 2001).In contrast, anosmia is characterized by a total loss of the abilityto smell. In both cases, there is damage in the olfactory area of thebrain affecting the individual's sense of smell. In a magazinearticle from USA Today, researchers at Stanford University statesthat "processing of either sniff or smell signals-air rushing up thenose or odorant molecules latching onto nerve cells-could bedefective in the many people who lose their ability to smell" (1998,December).

The experimental hypothesisexamines whether an individual with Alzheimer's disease will havemore olfactory brain activity than an individual with anosmia. TheSniffin' Sticks test battery will be used in combination with fMRI totest the olfactory activation levels of fifteen individuals withAlzheimer's disease and at least five individuals with anosmia. TheSniffin' Sticks test contains three different tests, odor threshold,odor discrimination, and odor identification (Schnieper,Welge-Luessen, & Wolfensberger, 2001). However, for the purposeof this study it will only be used in order to discriminate betweenthe two levels of the independent variable. The experiment containsone independent variable- odor, which consists of two levels,presence and absence. The dependent variable consists of the fMRIresults which determine the amount of oxygen in the blood in theolfactory bulb and can be rated on a scale from one to five (withfive being the most activation and one being the least activation).Another approach to this type of experiment would be to compare thehabituation rates and activation levels between the two types ofparticipants. Allin et al. (1997) found that after repeated olfactorystimulation, habituation occurred. A different approach to comparinganosmia and Alzheimer's disease would be to have the subjectsparticipate in the Güüttich test. Dresden et al.(1999)exposed their subjects to this type of smell test where "liquids areplaced on the participants' tongue and they are then asked todescribe the aroma of the liquid". The Güüttich test inthis experiment would not only provide a brain activation comparisonof smell but also one of taste between anosmics and Alzheimer'sparticipants.

Is there more olfactory activationin the brain when an odor is present? In one magazine article fromUSA Today, it was discovered that when sniffing in the absence of anodor, brain images showed that an area called the piriform cortex wasmost active. But "when they actually detected a smell, other areas inthe frontal lobe were most active" (1998, December). This informationallows one to see that an individual with Alzheimer's disease wouldhave more activation in the olfactory bulb than an anosmicindividual, because although a person with Alzheimer's disease hasolfactory damage, they still have a small ability to detect smell andthus receive stimulation from both sniffing and smelling. However,anosmic brains are only receiving stimulation from the sensation ofsniffing due to their total loss of ability to smell, and because ofthis loss will have less activation. Even though an odor may not bepresent, olfactory activation is still present when an individualsniffs. The problem at hand is determining whether olfactoryactivation is actually higher in individuals with Alzheimer's diseaseversus individuals with anosmia. Based on the above research it ishypothesized that individuals with Alzheimer's disease will have moreolfactory activation than an individual with generalanosmia.



Previous researchers studyinganosmia and Alzheimer's disease have often used a larger amount ofpeople with Alzheimer's compared to just one person with anosmia(Desmond et al., 1998). However, the results may be more significantif there were more than one anosmic participant. For this experimentit would be beneficial to use at least five anosmic individuals andat least fifteen individuals with Alzheimer's disease. Participantsshould be selected by either volunteering themselves or by randomsampling. Andrew et al. (2001) found that when using fMRI and testingyoung and elderly Alzheimer's participants, the older participantshad less brain activity. Thus, in order to receive more significantresults that apply to the entire population, this experiment shouldnot be age, gender, or race specific. Solicitation of the experimentcan include posting research opportunities in a newspaper, magazine,or even suggesting doctors to pass on the opportunity to Alzheimer'sor anosmic patients. If necessary, compensation for participating canbe provided; however, previous researchers have not compensated theirparticipants and have not had difficulty finding subjects. The onlyqualifications required for participation include a DSM-IV diagnosisof Alzheimer's disease or being anosmic.


The materials used in previousresearch included a fMRI or PET scan, a television screen viewed viaa mirror located in the fMRI machine, and a type of odoridentification test battery. In this experiment, the Sniffin' Stickstest battery will be used to manipulate the independent variable inorder to determine the difference in the levels of olfactoryactivation between Alzheimer's disease and anosmia . Unlike othersmell identification tests, the Sniffin' Sticks test is based onpen-like devices which contain different odors (Hummel, Kobal,Konnert, & Rossenheim, 2001). The importance of this test and howit will be used is described in the procedures section. Previousresearchers like Desmond et al. (1998) have used fMRI's to obtaintheir data. The fMRI is useful because it measures the amount ofoxygen in the blood in the area of the brain one is studying. In thisexperiment, the results from the fMRI will be obtained from thefrontal and temporal areas of the brain since these are the primaryolfactory areas. An image on a television screen will be reflectedonto a mirror located above the participants head in the fMRImachine, therefore, allowing the researchers to communicate with thesubjects during the experiment. For example, the researchers willflash the word "sniff" on the screen instructing the subject to sniffat that moment. Previous researchers using fMRI's have also usedthese mirrors and television screens to communicate with theirparticipants (Desmond et al., 1998).


After selection of theparticipants and upon visitation, subjects will be instructed to signa consent form and also provide adequate documentation of theirdiagnosis of either Alzheimer's disease or anosmia. Each subject willthen be informed of both the importance and purpose of the researchproject. The order of choosing an Alzheimer's subject or anosmicsubject is not important; however, it is necessary that thisinformation be recorded so that later, the researchers know whatbrain image corresponds to which type of participant. The participantwill then be taken to the room containing the fMRI machine where theexperiment will begin. Next, they will be placed into the fMRImachine and instructed to keep their eyes focused on the mirrorlocated above their head as to know when to sniff. The two levels ofthe independent variable will then be manipulated by the researchers.Whether the researcher starts with the presence or absence of theodor first is not important; however, it is necessary that thisinformation be recorded in order to know which fMRI image is due towhich level (presence versus absence). Next, the researcher willeither present/not present an odor using the Sniffin' Sticks test andretrieve the image of the activated areas of the brain. At this timeresearchers should be paying close attention to the different areasactivated by each participant (Desmond et al. ,1998). Eachparticipant should be subjected to at least five trials of each levelof the independent variable. For example, each subject shouldparticipate in at least five trials of an odor being present and atleast five trials of an odor being absent. Once the participant hasbeen exposed to ten trials, the researchers should then move onto thenext subject and repeat the process stated above. If the researchershave decided to compensate their subjects for participating in theexperiment, it should be offered at this time.


In this experiment there is onetrue independent variable, odor, and each participant is exposed toboth treatments (odor present and odor absent). The quasi-independentvariable is participant type: Alzheimers or a general anosmic. Thedependent variable consists of the fMRI results which determine theamount of oxygen in the blood that is present in the olfactory bulb.For these reasons, a two-factor mixed ANOVA may be used to analyzethe data. These results will then be applied to state whether theresearcher will reject or fail to reject the hypothesis. Aspreviously stated, the experimental hypothesis examines whether anindividual with Alzheimer's disease will have more olfactoryactivation than an individual with anosmia.



In this section, researchers willstart off by stating whether or not examination of the resultssupported the hypothesis. An example for this experiment may statethat examination of the results will support the hypothesis thatindividuals with Alzheimer's disease will have a higher level ofolfactory activation than an individual with anosmia. It is predictedthat the results of the ANOVA will reveal that there was asignificant difference between the ratings for odor presence and odorabsence in individuals with anosmia and Alzheimer's disease. Morespecifically, the individuals with Alzheimer's disease will havehigher olfactory activation levels compared to the individuals withanosmia. Looking again at the original hypothesis, it is evident thatthe proposed findings support the hypothesis.


The current experiment assessedthe difference between olfactory activation levels in anosmics andAlzheimer's disease participants based on whether an odor was presentor absent. According to the possible data, the level of activationwas affected by the presence/absence of an odor in each type ofparticipant.

There are many ways to replicatethis type of experiment for future research. For example, instead oftesting the activation levels for the presence/absence of an odor,researchers may try comparing anosmics and Alzheimer's subjects bymanipulating all three characteristics of the Sniffin' Sticks test(odor threshold, odor identification, and odor discrimination). Forinstance, when testing for odor discrimination participants could beasked to sniff three pens-two containing the same odor and onecontaining a different odor-and tell which one smelled different.Other groups of researchers have used a different type of smell testcalled the University of Pennsylvania Smell Identification Test(UPSIT). This quantitative test consists of forty odors where "eachodor is micro encapsulated on a pad so that the participant canscratch and sniff the odor" (Kondo et al., 1998).

Although the Sniffin' Sticks testwas used for this proposed experiment, as seen above there are manydifferent types of smell identification and discrimination testsavailable in order to replicate this experiment. Other ways toreplicate this experiment would be to test the olfactory activationlevels using a PET scan. In the various experiments used in thisproposal, almost all of the researchers used fMRI results. However,it is possible to achieve similar olfactory information using a PETscan. Changing the independent variable, the dependent variable andeven the type and amount of participants also allows for differentexperiments. Alzheimer's disease is not the only degenerative diseasethat is characterized by a decrease in the ability to smell. Otherdiseases such as Parkinson's and schizophrenia affect theindividual's ability to smell. Future researchers may also want toreplicate this experiment comparing schizophrenics and people withParkinson's disease. Wuensch (1993) states that more than two millionAmericans suffer from a significant loss of their ability to smell.With the information presented in this proposal and information yetto be discovered, researchers can apply their knowledge to futureexperiments in order to obtain a better understanding of thesediseases and help these people suffering from Alzheimer's,Parkinson's, schizophrenia and anosmia.



Allin, Andrew, Brammer, Bullmore, Doty, & Howard et al. (1997).Functional MR imaging during odor stimulation: Preliminary Data,204(3), 833-838.

Andrew, C., Bryant, C., Critchley,H.D., Fukuda, R., Howard, R., & Jackson, S.H.D. et al. (2001).Functional magnetic resonance imaging of odor identification: Theeffect of aging. Journals of Gerontology: Series A,56A(12), M756-M760.

Desmond, J.E., Gabrieli, J.D.E.,Glover, G.H., Goode, R.L., Prabhakaran, V., & Sobel, N. et al.(1998). Sniffing and smelling: Separate subsystems in the humanolfactory cortex. Nature, 392(6673), 282-286.

Dresden, Fetscherstr, Heilmann,Hummel, Huuttenbrink, & Knecht et al. (1999). Retronasalolfactory function: clinical use of the "Güüttich"-test inthe diagnostics of olfactory disorders, 78, 627-631.

Gabrielli, J. (1998, December).Sniffing may prepare brain for smelling. USA Today, 127(2643),15.

Hummel, Kobal, Konnert, &Rossenheim (2001). Screening of Olfactory Function using a 3minute odor identification test: reliability, normative data, andinvestigations in patients with olfactory loss [Online].Available:

Hummel, Kobal, Pauli, Sekinger,& Wolf (1997). Sniffin' Sticks": Olfactory performance assessedby the combined testing of odor identification, odor discriminationand olfactory threshold, 22(1), 39-52.

Hummel, Moll, Quint, Temmel,Welge-Luessen, & Wolf (2001). Olfactory function in patients witholfactory groove meningioma, 70(2), 218.

Kondo et al. (1998). A study ofthe relationship between the T&T olfactometer and the Universityof Pennsylvania Smell Identification Test in a Japanese Population,12(5), 353-358. [Online]

Schnieper, I., Welge-Lussen, A.,& Wolfensberger, M. (2001). Sniffin' Sticks: A New Olfactory TestBattery. Acta Oto-LaryngologicaJournal,120(2), 303-307.

Wuensch, K.L. (2001, November 10).How Frequent is Anosmia? [Online]. Available: